Alex Zhouk

{
  "first_name": "Alex",
  "last_name": "Zhouk",
  "email": "{email}",
  "phone": "{phone}",
  "country_short_name": "US",
  "resume_text": "ALEX ZHOUK\nSeattle, WA | https://www.linkedin.com/in/alexzhouk/ | https://kvkeys.com/#sql | https://kvmachines.com\nUS citizen. Based in Seattle. Open to Washington, California, and New York. Available to start on offer.\n\nSUMMARY\nSenior data engineer at Coinbase on dbt, Airflow and Snowflake: the unified staking data model, a semantic layer made LLM-ready with the Claude API, the observability platform behind 20+ KPIs. Data Engineer titles since 2023 after a Data Scientist role at TikTok and BI and analytics roles at Amazon; 10+ years in data roles. This year, KVMACHINES: an independent project to test whether agents can run a business on Anthropic's products, with a person's signature on every step that changes something (kvkeys.com/#sql).\n\nSKILLS\nPipelines: dbt Core and Cloud, Airflow, Spark and PySpark, Hive, Databricks, AWS Glue and EMR, GitHub Actions CI/CD\nWarehouses: Snowflake, Redshift, Databricks, Azure Data Lake, Postgres; dimensional modeling (star schemas, slowly changing dimensions), semantic layers, data contracts, freshness SLAs and data quality KPIs\nLanguages: SQL, Python (Pandas, PySpark, FastAPI, Streamlit), R\nReporting: Superset, Tableau, Streamlit apps, SQL consoles in the browser (Postgres in the page); executive and cross-functional dashboards\nAI-assisted engineering: Claude Code, the Claude API and MCP connectors for pipeline, schema and CI work, with an independent review session before the merge; LLM-ready dbt metadata\nCloud and tooling: AWS, Azure, Docker, GitHub Actions\n\nEXPERIENCE\n\nSenior Data Engineer\nCoinbase | Remote | Mar 2024 - Present\n- Led the unified staking data model, a dimensional model across $18.9B in crypto assets on six networks, with Product Management, Data Science and leadership, using dbt Core and Cloud, Airflow and Snowflake; executive dashboards in Superset and Tableau.\n- Standardized the semantic layer across Coinbase: Python automation on the Claude API converted 1,600+ dbt, YAML and SQL files into LLM-ready metadata in the company's first data contracts rollout, so LLM-based analytics tools query trusted, documented datasets directly.\n- Built the org-wide observability platform in Streamlit (Python, SQL) that monitors 20+ pipeline health and data quality KPIs, with automated source freshness tests and schema validation across the warehouse, cutting the time to detect a failure.\n\nSenior Data Engineer\nMachina Labs | Chatsworth, CA | Feb 2023 - Mar 2024\n- First data engineering hire: built a robotics startup's Databricks warehouse from the ground up, PySpark ELT pipelines on Azure that bring robot telemetry, Salesforce and GraphQL feeds into a medallion lake for 40+ stakeholders, with CI/CD through GitHub Actions.\n- Built the FastAPI service over the Databricks APIs that ended direct cluster access for three internal teams; shipped the invoice management app and a bidirectional pricing calculator (Streamlit, PySpark, Docker on Azure) with write-back for Finance and Sales.\n\nData Scientist, Creator Monetization\nTikTok | Los Angeles, CA | Dec 2021 - Feb 2023\n- Owned the Creator Monetization data stack end to end: 15+ production tables with dimensional modeling and 50+ Apache Hive SQL pipelines behind dashboards and cross-team analytics, with data contracts between event producers and downstream consumers.\n- Found and fixed a systematic error in the Branded Content revenue pipeline through SQL and lineage review, correcting projected revenue by $115M (+10%); built the Instacart add-to-cart ETL and reporting with anomaly detection, alerting and source freshness SLAs.\n\nBusiness Intelligence Engineer II, Sponsored Products Advertising\nAmazon | Seattle, WA | May 2020 - Dec 2021\n- Built and owned four Redshift ETL pipelines (80+ tasks) that publish the Weekly Business Review for the Targeting, Optimization, Recommendation and Budget teams: $450M+ in weekly ad revenue, zero SLA misses in 18 months; seven Tableau dashboards.\n- Owned the A/B test that launched Thematic placement sourcing (+$65MM annualized) and identified $8MM more in annualized revenue; standardized data models adopted across four teams.\n\nBusiness Intelligence Engineer II, Private Brands\nAmazon | Seattle, WA | Nov 2018 - May 2020\n- Automated product metrics in Redshift for the Accelerator program across 11 countries ($800M, 800% year over year), with slowly changing dimension tables for product attribute history; built the placement-level sales dashboard that steered marketing spend.\n- Traced sudden sales dips to bad actors ($300k opportunity cost) and shipped a real-time monitoring pipeline with data quality alerting and the takedown dashboard.\n\nCustom Pricing Analyst\nAmazon Web Services | Seattle, WA | Oct 2017 - Nov 2018\n- Designed and built the Enterprise Support analytical warehouse in Redshift (10M+ rows), a star-schema dimensional model that replaced manual billing workflows.\n- Owned the monthly billing pipeline for every AWS Enterprise Support customer: $55M a month in recurring billing, zero disputes; led two projects that added $1M in monthly AWS revenue.\n\nBusiness Analyst (contract)\nAmazon | Seattle, WA | Dec 2016 - Oct 2017\n- Built the Redshift pipelines and dashboards for contract-worker onboarding across 40+ countries and set the data governance standards for the HR system integrations; SQL, R and Python tooling for job-code mapping of the global contingent workforce.\n\nCompensation Data Analyst (contract)\nMicrosoft | Redmond, WA | Sep 2016 - Dec 2016\n- R scripts that automated the manual revenue adjustment and bulk territory processes (60+ hours saved a month); quarterly audits of revenue, territories and quota.\n\nFinancial Analyst\nBooking.com | Bellevue, WA | Jul 2015 - Sep 2015\n- Consolidated invoice reporting in SQL and Excel across two offices.\n\nFinancial Analyst, Retail\nAmazon | Seattle, WA | Sep 2014 - Mar 2015\n- Optimized the SQL and automated the refresh of 20 Excel reports behind the Canada Retail weekly business review; customer returns and free replacements analyzed by reason code.\n\nPROJECTS\n\nKVMACHINES, independent project\nkvkeys.com/#sql, kvmachines.com | Seattle, WA | Jan 2026 - Present\n- Built to test whether agents can run a business on Anthropic's products: one Claude session per department with its own key, memory store and routines; a person's signature on every step that changes something; every output a record people and programs can check.\n- Designed the kvkeys data model: every connector, skill and version of the department plugins in one Postgres schema (818 entities, 351 skills, 420 versions), replayed from history, every table and most columns commented, a 20-step CI gate before any merge.\n- Shipped kvkeys.com/#sql, Postgres running in the browser as the self-serve SQL console over that schema; built the jobs dataset behind this application: 293,891 postings from 13,750 Greenhouse, Lever and Ashby boards joined to the BLS occupation and employment tables.\n- Agents open the pull requests and a separate session reviews them before the merge, under rules I set: 685 merged across 39 repositories from 2 to 13 October 2026.\n\nEDUCATION\nUniversity of Washington | Seattle, WA\nBachelor of Arts, Economics; Minor, Entrepreneurship | Sep 2011 - Dec 2015\nGPA 3.5, Dean's List Scholar\n",
  "employments": [
    {
      "company_name": "Coinbase",
      "title": "Senior Data Engineer",
      "start_date": {
        "month": "3",
        "year": "2024"
      },
      "current": "true"
    },
    {
      "company_name": "Machina Labs",
      "title": "Senior Data Engineer",
      "start_date": {
        "month": "2",
        "year": "2023"
      },
      "end_date": {
        "month": "3",
        "year": "2024"
      },
      "current": "false"
    },
    {
      "company_name": "TikTok",
      "title": "Data Scientist, Creator Monetization",
      "start_date": {
        "month": "12",
        "year": "2021"
      },
      "end_date": {
        "month": "2",
        "year": "2023"
      },
      "current": "false"
    },
    {
      "company_name": "Amazon",
      "title": "Business Intelligence Engineer II, Sponsored Products Advertising",
      "start_date": {
        "month": "5",
        "year": "2020"
      },
      "end_date": {
        "month": "12",
        "year": "2021"
      },
      "current": "false"
    },
    {
      "company_name": "Amazon",
      "title": "Business Intelligence Engineer II, Private Brands",
      "start_date": {
        "month": "11",
        "year": "2018"
      },
      "end_date": {
        "month": "5",
        "year": "2020"
      },
      "current": "false"
    },
    {
      "company_name": "Amazon Web Services",
      "title": "Custom Pricing Analyst",
      "start_date": {
        "month": "10",
        "year": "2017"
      },
      "end_date": {
        "month": "11",
        "year": "2018"
      },
      "current": "false"
    },
    {
      "company_name": "Amazon",
      "title": "Business Analyst (contract)",
      "start_date": {
        "month": "12",
        "year": "2016"
      },
      "end_date": {
        "month": "10",
        "year": "2017"
      },
      "current": "false"
    },
    {
      "company_name": "Microsoft",
      "title": "Compensation Data Analyst (contract)",
      "start_date": {
        "month": "9",
        "year": "2016"
      },
      "end_date": {
        "month": "12",
        "year": "2016"
      },
      "current": "false"
    },
    {
      "company_name": "Booking.com",
      "title": "Financial Analyst",
      "start_date": {
        "month": "7",
        "year": "2015"
      },
      "end_date": {
        "month": "9",
        "year": "2015"
      },
      "current": "false"
    },
    {
      "company_name": "Amazon",
      "title": "Financial Analyst, Retail",
      "start_date": {
        "month": "9",
        "year": "2014"
      },
      "end_date": {
        "month": "3",
        "year": "2015"
      },
      "current": "false"
    }
  ],
  "educations": [
    {
      "school_name_id": {
        "lookup": "GET /v1/boards/{board_token}/education/schools?term=University%20of%20Washington",
        "pick": "items[].id where text is \"University of Washington\"",
        "doc": "_educations.md"
      },
      "degree_id": {
        "lookup": "GET /v1/boards/{board_token}/education/degrees?term=Bachelor's%20Degree",
        "pick": "items[].id where text is \"Bachelor's Degree\"",
        "doc": "_educations.md"
      },
      "discipline_id": {
        "lookup": "GET /v1/boards/{board_token}/education/disciplines?term=Economics",
        "pick": "items[].id where text is \"Economics\"",
        "doc": "_educations.md"
      },
      "start_date": {
        "month": "9",
        "year": "2011"
      },
      "end_date": {
        "month": "12",
        "year": "2015"
      }
    }
  ]
}

{
  "id": null,
  "title": "Senior Data Engineer",
  "company_name": "Alex Zhouk",
  "first_published": "2026-10-13T00:00:00Z",
  "updated_at": "2026-10-13T00:00:00Z",
  "application_deadline": null,
  "requisition_id": "resume v12",
  "location": {
    "name": "Seattle, WA"
  },
  "content": "<p>Senior data engineer at Coinbase on dbt, Airflow and Snowflake: the unified staking data model, a semantic layer made LLM-ready with the Claude API, the observability platform behind 20+ KPIs. Data Engineer titles since 2023 after a Data Scientist role at TikTok and BI and analytics roles at Amazon; 10+ years in data roles. This year, KVMACHINES: an independent project to test whether agents can run a business on Anthropic's products, with a person's signature on every step that changes something (kvkeys.com/#sql).</p><h3>Skills</h3><ul><li><strong>Pipelines:</strong> dbt Core and Cloud, Airflow, Spark and PySpark, Hive, Databricks, AWS Glue and EMR, GitHub Actions CI/CD</li><li><strong>Warehouses:</strong> Snowflake, Redshift, Databricks, Azure Data Lake, Postgres; dimensional modeling (star schemas, slowly changing dimensions), semantic layers, data contracts, freshness SLAs and data quality KPIs</li><li><strong>Languages:</strong> SQL, Python (Pandas, PySpark, FastAPI, Streamlit), R</li><li><strong>Reporting:</strong> Superset, Tableau, Streamlit apps, SQL consoles in the browser (Postgres in the page); executive and cross-functional dashboards</li><li><strong>AI-assisted engineering:</strong> Claude Code, the Claude API and MCP connectors for pipeline, schema and CI work, with an independent review session before the merge; LLM-ready dbt metadata</li><li><strong>Cloud and tooling:</strong> AWS, Azure, Docker, GitHub Actions</li></ul><h3>Experience</h3><ul><li><strong>Senior Data Engineer</strong>, Coinbase (Remote, Mar 2024 - Present)<ul><li>Led the unified staking data model, a dimensional model across $18.9B in crypto assets on six networks, with Product Management, Data Science and leadership, using dbt Core and Cloud, Airflow and Snowflake; executive dashboards in Superset and Tableau.</li><li>Standardized the semantic layer across Coinbase: Python automation on the Claude API converted 1,600+ dbt, YAML and SQL files into LLM-ready metadata in the company's first data contracts rollout, so LLM-based analytics tools query trusted, documented datasets directly.</li><li>Built the org-wide observability platform in Streamlit (Python, SQL) that monitors 20+ pipeline health and data quality KPIs, with automated source freshness tests and schema validation across the warehouse, cutting the time to detect a failure.</li></ul></li><li><strong>Senior Data Engineer</strong>, Machina Labs (Chatsworth, CA, Feb 2023 - Mar 2024)<ul><li>First data engineering hire: built a robotics startup's Databricks warehouse from the ground up, PySpark ELT pipelines on Azure that bring robot telemetry, Salesforce and GraphQL feeds into a medallion lake for 40+ stakeholders, with CI/CD through GitHub Actions.</li><li>Built the FastAPI service over the Databricks APIs that ended direct cluster access for three internal teams; shipped the invoice management app and a bidirectional pricing calculator (Streamlit, PySpark, Docker on Azure) with write-back for Finance and Sales.</li></ul></li><li><strong>Data Scientist, Creator Monetization</strong>, TikTok (Los Angeles, CA, Dec 2021 - Feb 2023)<ul><li>Owned the Creator Monetization data stack end to end: 15+ production tables with dimensional modeling and 50+ Apache Hive SQL pipelines behind dashboards and cross-team analytics, with data contracts between event producers and downstream consumers.</li><li>Found and fixed a systematic error in the Branded Content revenue pipeline through SQL and lineage review, correcting projected revenue by $115M (+10%); built the Instacart add-to-cart ETL and reporting with anomaly detection, alerting and source freshness SLAs.</li></ul></li><li><strong>Business Intelligence Engineer II, Sponsored Products Advertising</strong>, Amazon (Seattle, WA, May 2020 - Dec 2021)<ul><li>Built and owned four Redshift ETL pipelines (80+ tasks) that publish the Weekly Business Review for the Targeting, Optimization, Recommendation and Budget teams: $450M+ in weekly ad revenue, zero SLA misses in 18 months; seven Tableau dashboards.</li><li>Owned the A/B test that launched Thematic placement sourcing (+$65MM annualized) and identified $8MM more in annualized revenue; standardized data models adopted across four teams.</li></ul></li><li><strong>Business Intelligence Engineer II, Private Brands</strong>, Amazon (Seattle, WA, Nov 2018 - May 2020)<ul><li>Automated product metrics in Redshift for the Accelerator program across 11 countries ($800M, 800% year over year), with slowly changing dimension tables for product attribute history; built the placement-level sales dashboard that steered marketing spend.</li><li>Traced sudden sales dips to bad actors ($300k opportunity cost) and shipped a real-time monitoring pipeline with data quality alerting and the takedown dashboard.</li></ul></li><li><strong>Custom Pricing Analyst</strong>, Amazon Web Services (Seattle, WA, Oct 2017 - Nov 2018)<ul><li>Designed and built the Enterprise Support analytical warehouse in Redshift (10M+ rows), a star-schema dimensional model that replaced manual billing workflows.</li><li>Owned the monthly billing pipeline for every AWS Enterprise Support customer: $55M a month in recurring billing, zero disputes; led two projects that added $1M in monthly AWS revenue.</li></ul></li><li><strong>Business Analyst (contract)</strong>, Amazon (Seattle, WA, Dec 2016 - Oct 2017)<ul><li>Built the Redshift pipelines and dashboards for contract-worker onboarding across 40+ countries and set the data governance standards for the HR system integrations; SQL, R and Python tooling for job-code mapping of the global contingent workforce.</li></ul></li><li><strong>Compensation Data Analyst (contract)</strong>, Microsoft (Redmond, WA, Sep 2016 - Dec 2016)<ul><li>R scripts that automated the manual revenue adjustment and bulk territory processes (60+ hours saved a month); quarterly audits of revenue, territories and quota.</li></ul></li><li><strong>Financial Analyst</strong>, Booking.com (Bellevue, WA, Jul 2015 - Sep 2015)<ul><li>Consolidated invoice reporting in SQL and Excel across two offices.</li></ul></li><li><strong>Financial Analyst, Retail</strong>, Amazon (Seattle, WA, Sep 2014 - Mar 2015)<ul><li>Optimized the SQL and automated the refresh of 20 Excel reports behind the Canada Retail weekly business review; customer returns and free replacements analyzed by reason code.</li></ul></li></ul><h3>Projects</h3><ul><li><strong>KVMACHINES, independent project</strong>, kvkeys.com/#sql, kvmachines.com (Seattle, WA, Jan 2026 - Present)<ul><li>Built to test whether agents can run a business on Anthropic's products: one Claude session per department with its own key, memory store and routines; a person's signature on every step that changes something; every output a record people and programs can check.</li><li>Designed the kvkeys data model: every connector, skill and version of the department plugins in one Postgres schema (818 entities, 351 skills, 420 versions), replayed from history, every table and most columns commented, a 20-step CI gate before any merge.</li><li>Shipped kvkeys.com/#sql, Postgres running in the browser as the self-serve SQL console over that schema; built the jobs dataset behind this application: 293,891 postings from 13,750 Greenhouse, Lever and Ashby boards joined to the BLS occupation and employment tables.</li><li>Agents open the pull requests and a separate session reviews them before the merge, under rules I set: 685 merged across 39 repositories from 2 to 13 October 2026.</li></ul></li></ul><h3>Education</h3><ul><li><strong>University of Washington</strong>, Seattle, WA: Bachelor of Arts, Economics; Minor, Entrepreneurship (Sep 2011 - Dec 2015); GPA 3.5, Dean's List Scholar</li></ul><p>US citizen. Based in Seattle. Open to Washington, California, and New York. Available to start on offer.</p>",
  "absolute_url": "https://alexzhouk.com/",
  "language": "en",
  "internal_job_id": null,
  "metadata": [
    {
      "id": null,
      "name": "Status",
      "value_type": "short_text",
      "value": "US citizen. Based in Seattle. Open to Washington, California, and New York. Available to start on offer."
    },
    {
      "id": null,
      "name": "LinkedIn",
      "value_type": "url",
      "value": "https://www.linkedin.com/in/alexzhouk/"
    },
    {
      "id": null,
      "name": "kvkeys.com/#sql",
      "value_type": "url",
      "value": "https://kvkeys.com/#sql"
    },
    {
      "id": null,
      "name": "kvmachines.com",
      "value_type": "url",
      "value": "https://kvmachines.com"
    }
  ],
  "questions": [
    {
      "required": true,
      "label": "First Name",
      "fields": [
        {
          "name": "first_name",
          "type": "input_text"
        }
      ]
    },
    {
      "required": true,
      "label": "Last Name",
      "fields": [
        {
          "name": "last_name",
          "type": "input_text"
        }
      ]
    },
    {
      "required": true,
      "label": "Email",
      "fields": [
        {
          "name": "email",
          "type": "input_text"
        }
      ]
    },
    {
      "required": false,
      "label": "Phone",
      "fields": [
        {
          "name": "phone",
          "type": "input_text"
        }
      ]
    },
    {
      "required": true,
      "label": "Resume",
      "fields": [
        {
          "name": "resume",
          "type": "input_file"
        },
        {
          "name": "resume_text",
          "type": "textarea"
        }
      ]
    }
  ]
}

{
  "name": "Alex Zhouk",
  "email": "{email}",
  "phone": "{phone}",
  "org": "Coinbase",
  "urls": {
    "LinkedIn": "https://www.linkedin.com/in/alexzhouk/",
    "Portfolio": "https://kvkeys.com/#sql",
    "Other": "https://kvmachines.com"
  },
  "comments": "US citizen. Based in Seattle. Open to Washington, California, and New York. Available to start on offer."
}

{
  "id": "alexzhouk-resume-v12",
  "text": "Senior Data Engineer",
  "categories": {
    "location": "Seattle, WA",
    "commitment": null,
    "team": null,
    "department": null,
    "allLocations": [
      "Seattle, WA",
      "Washington",
      "California",
      "New York"
    ]
  },
  "country": "US",
  "opening": "<p>Senior data engineer at Coinbase on dbt, Airflow and Snowflake: the unified staking data model, a semantic layer made LLM-ready with the Claude API, the observability platform behind 20+ KPIs. Data Engineer titles since 2023 after a Data Scientist role at TikTok and BI and analytics roles at Amazon; 10+ years in data roles. This year, KVMACHINES: an independent project to test whether agents can run a business on Anthropic's products, with a person's signature on every step that changes something (kvkeys.com/#sql).</p>",
  "openingPlain": "Senior data engineer at Coinbase on dbt, Airflow and Snowflake: the unified staking data model, a semantic layer made LLM-ready with the Claude API, the observability platform behind 20+ KPIs. Data Engineer titles since 2023 after a Data Scientist role at TikTok and BI and analytics roles at Amazon; 10+ years in data roles. This year, KVMACHINES: an independent project to test whether agents can run a business on Anthropic's products, with a person's signature on every step that changes something (kvkeys.com/#sql).",
  "description": "<p>Senior data engineer at Coinbase on dbt, Airflow and Snowflake: the unified staking data model, a semantic layer made LLM-ready with the Claude API, the observability platform behind 20+ KPIs. Data Engineer titles since 2023 after a Data Scientist role at TikTok and BI and analytics roles at Amazon; 10+ years in data roles. This year, KVMACHINES: an independent project to test whether agents can run a business on Anthropic's products, with a person's signature on every step that changes something (kvkeys.com/#sql).</p><h3>Skills</h3><ul><li><strong>Pipelines:</strong> dbt Core and Cloud, Airflow, Spark and PySpark, Hive, Databricks, AWS Glue and EMR, GitHub Actions CI/CD</li><li><strong>Warehouses:</strong> Snowflake, Redshift, Databricks, Azure Data Lake, Postgres; dimensional modeling (star schemas, slowly changing dimensions), semantic layers, data contracts, freshness SLAs and data quality KPIs</li><li><strong>Languages:</strong> SQL, Python (Pandas, PySpark, FastAPI, Streamlit), R</li><li><strong>Reporting:</strong> Superset, Tableau, Streamlit apps, SQL consoles in the browser (Postgres in the page); executive and cross-functional dashboards</li><li><strong>AI-assisted engineering:</strong> Claude Code, the Claude API and MCP connectors for pipeline, schema and CI work, with an independent review session before the merge; LLM-ready dbt metadata</li><li><strong>Cloud and tooling:</strong> AWS, Azure, Docker, GitHub Actions</li></ul><h3>Experience</h3><ul><li><strong>Senior Data Engineer</strong>, Coinbase (Remote, Mar 2024 - Present)<ul><li>Led the unified staking data model, a dimensional model across $18.9B in crypto assets on six networks, with Product Management, Data Science and leadership, using dbt Core and Cloud, Airflow and Snowflake; executive dashboards in Superset and Tableau.</li><li>Standardized the semantic layer across Coinbase: Python automation on the Claude API converted 1,600+ dbt, YAML and SQL files into LLM-ready metadata in the company's first data contracts rollout, so LLM-based analytics tools query trusted, documented datasets directly.</li><li>Built the org-wide observability platform in Streamlit (Python, SQL) that monitors 20+ pipeline health and data quality KPIs, with automated source freshness tests and schema validation across the warehouse, cutting the time to detect a failure.</li></ul></li><li><strong>Senior Data Engineer</strong>, Machina Labs (Chatsworth, CA, Feb 2023 - Mar 2024)<ul><li>First data engineering hire: built a robotics startup's Databricks warehouse from the ground up, PySpark ELT pipelines on Azure that bring robot telemetry, Salesforce and GraphQL feeds into a medallion lake for 40+ stakeholders, with CI/CD through GitHub Actions.</li><li>Built the FastAPI service over the Databricks APIs that ended direct cluster access for three internal teams; shipped the invoice management app and a bidirectional pricing calculator (Streamlit, PySpark, Docker on Azure) with write-back for Finance and Sales.</li></ul></li><li><strong>Data Scientist, Creator Monetization</strong>, TikTok (Los Angeles, CA, Dec 2021 - Feb 2023)<ul><li>Owned the Creator Monetization data stack end to end: 15+ production tables with dimensional modeling and 50+ Apache Hive SQL pipelines behind dashboards and cross-team analytics, with data contracts between event producers and downstream consumers.</li><li>Found and fixed a systematic error in the Branded Content revenue pipeline through SQL and lineage review, correcting projected revenue by $115M (+10%); built the Instacart add-to-cart ETL and reporting with anomaly detection, alerting and source freshness SLAs.</li></ul></li><li><strong>Business Intelligence Engineer II, Sponsored Products Advertising</strong>, Amazon (Seattle, WA, May 2020 - Dec 2021)<ul><li>Built and owned four Redshift ETL pipelines (80+ tasks) that publish the Weekly Business Review for the Targeting, Optimization, Recommendation and Budget teams: $450M+ in weekly ad revenue, zero SLA misses in 18 months; seven Tableau dashboards.</li><li>Owned the A/B test that launched Thematic placement sourcing (+$65MM annualized) and identified $8MM more in annualized revenue; standardized data models adopted across four teams.</li></ul></li><li><strong>Business Intelligence Engineer II, Private Brands</strong>, Amazon (Seattle, WA, Nov 2018 - May 2020)<ul><li>Automated product metrics in Redshift for the Accelerator program across 11 countries ($800M, 800% year over year), with slowly changing dimension tables for product attribute history; built the placement-level sales dashboard that steered marketing spend.</li><li>Traced sudden sales dips to bad actors ($300k opportunity cost) and shipped a real-time monitoring pipeline with data quality alerting and the takedown dashboard.</li></ul></li><li><strong>Custom Pricing Analyst</strong>, Amazon Web Services (Seattle, WA, Oct 2017 - Nov 2018)<ul><li>Designed and built the Enterprise Support analytical warehouse in Redshift (10M+ rows), a star-schema dimensional model that replaced manual billing workflows.</li><li>Owned the monthly billing pipeline for every AWS Enterprise Support customer: $55M a month in recurring billing, zero disputes; led two projects that added $1M in monthly AWS revenue.</li></ul></li><li><strong>Business Analyst (contract)</strong>, Amazon (Seattle, WA, Dec 2016 - Oct 2017)<ul><li>Built the Redshift pipelines and dashboards for contract-worker onboarding across 40+ countries and set the data governance standards for the HR system integrations; SQL, R and Python tooling for job-code mapping of the global contingent workforce.</li></ul></li><li><strong>Compensation Data Analyst (contract)</strong>, Microsoft (Redmond, WA, Sep 2016 - Dec 2016)<ul><li>R scripts that automated the manual revenue adjustment and bulk territory processes (60+ hours saved a month); quarterly audits of revenue, territories and quota.</li></ul></li><li><strong>Financial Analyst</strong>, Booking.com (Bellevue, WA, Jul 2015 - Sep 2015)<ul><li>Consolidated invoice reporting in SQL and Excel across two offices.</li></ul></li><li><strong>Financial Analyst, Retail</strong>, Amazon (Seattle, WA, Sep 2014 - Mar 2015)<ul><li>Optimized the SQL and automated the refresh of 20 Excel reports behind the Canada Retail weekly business review; customer returns and free replacements analyzed by reason code.</li></ul></li></ul><h3>Projects</h3><ul><li><strong>KVMACHINES, independent project</strong>, kvkeys.com/#sql, kvmachines.com (Seattle, WA, Jan 2026 - Present)<ul><li>Built to test whether agents can run a business on Anthropic's products: one Claude session per department with its own key, memory store and routines; a person's signature on every step that changes something; every output a record people and programs can check.</li><li>Designed the kvkeys data model: every connector, skill and version of the department plugins in one Postgres schema (818 entities, 351 skills, 420 versions), replayed from history, every table and most columns commented, a 20-step CI gate before any merge.</li><li>Shipped kvkeys.com/#sql, Postgres running in the browser as the self-serve SQL console over that schema; built the jobs dataset behind this application: 293,891 postings from 13,750 Greenhouse, Lever and Ashby boards joined to the BLS occupation and employment tables.</li><li>Agents open the pull requests and a separate session reviews them before the merge, under rules I set: 685 merged across 39 repositories from 2 to 13 October 2026.</li></ul></li></ul><h3>Education</h3><ul><li><strong>University of Washington</strong>, Seattle, WA: Bachelor of Arts, Economics; Minor, Entrepreneurship (Sep 2011 - Dec 2015); GPA 3.5, Dean's List Scholar</li></ul>",
  "descriptionPlain": "ALEX ZHOUK\nSeattle, WA | https://www.linkedin.com/in/alexzhouk/ | https://kvkeys.com/#sql | https://kvmachines.com\nUS citizen. Based in Seattle. Open to Washington, California, and New York. Available to start on offer.\n\nSUMMARY\nSenior data engineer at Coinbase on dbt, Airflow and Snowflake: the unified staking data model, a semantic layer made LLM-ready with the Claude API, the observability platform behind 20+ KPIs. Data Engineer titles since 2023 after a Data Scientist role at TikTok and BI and analytics roles at Amazon; 10+ years in data roles. This year, KVMACHINES: an independent project to test whether agents can run a business on Anthropic's products, with a person's signature on every step that changes something (kvkeys.com/#sql).\n\nSKILLS\nPipelines: dbt Core and Cloud, Airflow, Spark and PySpark, Hive, Databricks, AWS Glue and EMR, GitHub Actions CI/CD\nWarehouses: Snowflake, Redshift, Databricks, Azure Data Lake, Postgres; dimensional modeling (star schemas, slowly changing dimensions), semantic layers, data contracts, freshness SLAs and data quality KPIs\nLanguages: SQL, Python (Pandas, PySpark, FastAPI, Streamlit), R\nReporting: Superset, Tableau, Streamlit apps, SQL consoles in the browser (Postgres in the page); executive and cross-functional dashboards\nAI-assisted engineering: Claude Code, the Claude API and MCP connectors for pipeline, schema and CI work, with an independent review session before the merge; LLM-ready dbt metadata\nCloud and tooling: AWS, Azure, Docker, GitHub Actions\n\nEXPERIENCE\n\nSenior Data Engineer\nCoinbase | Remote | Mar 2024 - Present\n- Led the unified staking data model, a dimensional model across $18.9B in crypto assets on six networks, with Product Management, Data Science and leadership, using dbt Core and Cloud, Airflow and Snowflake; executive dashboards in Superset and Tableau.\n- Standardized the semantic layer across Coinbase: Python automation on the Claude API converted 1,600+ dbt, YAML and SQL files into LLM-ready metadata in the company's first data contracts rollout, so LLM-based analytics tools query trusted, documented datasets directly.\n- Built the org-wide observability platform in Streamlit (Python, SQL) that monitors 20+ pipeline health and data quality KPIs, with automated source freshness tests and schema validation across the warehouse, cutting the time to detect a failure.\n\nSenior Data Engineer\nMachina Labs | Chatsworth, CA | Feb 2023 - Mar 2024\n- First data engineering hire: built a robotics startup's Databricks warehouse from the ground up, PySpark ELT pipelines on Azure that bring robot telemetry, Salesforce and GraphQL feeds into a medallion lake for 40+ stakeholders, with CI/CD through GitHub Actions.\n- Built the FastAPI service over the Databricks APIs that ended direct cluster access for three internal teams; shipped the invoice management app and a bidirectional pricing calculator (Streamlit, PySpark, Docker on Azure) with write-back for Finance and Sales.\n\nData Scientist, Creator Monetization\nTikTok | Los Angeles, CA | Dec 2021 - Feb 2023\n- Owned the Creator Monetization data stack end to end: 15+ production tables with dimensional modeling and 50+ Apache Hive SQL pipelines behind dashboards and cross-team analytics, with data contracts between event producers and downstream consumers.\n- Found and fixed a systematic error in the Branded Content revenue pipeline through SQL and lineage review, correcting projected revenue by $115M (+10%); built the Instacart add-to-cart ETL and reporting with anomaly detection, alerting and source freshness SLAs.\n\nBusiness Intelligence Engineer II, Sponsored Products Advertising\nAmazon | Seattle, WA | May 2020 - Dec 2021\n- Built and owned four Redshift ETL pipelines (80+ tasks) that publish the Weekly Business Review for the Targeting, Optimization, Recommendation and Budget teams: $450M+ in weekly ad revenue, zero SLA misses in 18 months; seven Tableau dashboards.\n- Owned the A/B test that launched Thematic placement sourcing (+$65MM annualized) and identified $8MM more in annualized revenue; standardized data models adopted across four teams.\n\nBusiness Intelligence Engineer II, Private Brands\nAmazon | Seattle, WA | Nov 2018 - May 2020\n- Automated product metrics in Redshift for the Accelerator program across 11 countries ($800M, 800% year over year), with slowly changing dimension tables for product attribute history; built the placement-level sales dashboard that steered marketing spend.\n- Traced sudden sales dips to bad actors ($300k opportunity cost) and shipped a real-time monitoring pipeline with data quality alerting and the takedown dashboard.\n\nCustom Pricing Analyst\nAmazon Web Services | Seattle, WA | Oct 2017 - Nov 2018\n- Designed and built the Enterprise Support analytical warehouse in Redshift (10M+ rows), a star-schema dimensional model that replaced manual billing workflows.\n- Owned the monthly billing pipeline for every AWS Enterprise Support customer: $55M a month in recurring billing, zero disputes; led two projects that added $1M in monthly AWS revenue.\n\nBusiness Analyst (contract)\nAmazon | Seattle, WA | Dec 2016 - Oct 2017\n- Built the Redshift pipelines and dashboards for contract-worker onboarding across 40+ countries and set the data governance standards for the HR system integrations; SQL, R and Python tooling for job-code mapping of the global contingent workforce.\n\nCompensation Data Analyst (contract)\nMicrosoft | Redmond, WA | Sep 2016 - Dec 2016\n- R scripts that automated the manual revenue adjustment and bulk territory processes (60+ hours saved a month); quarterly audits of revenue, territories and quota.\n\nFinancial Analyst\nBooking.com | Bellevue, WA | Jul 2015 - Sep 2015\n- Consolidated invoice reporting in SQL and Excel across two offices.\n\nFinancial Analyst, Retail\nAmazon | Seattle, WA | Sep 2014 - Mar 2015\n- Optimized the SQL and automated the refresh of 20 Excel reports behind the Canada Retail weekly business review; customer returns and free replacements analyzed by reason code.\n\nPROJECTS\n\nKVMACHINES, independent project\nkvkeys.com/#sql, kvmachines.com | Seattle, WA | Jan 2026 - Present\n- Built to test whether agents can run a business on Anthropic's products: one Claude session per department with its own key, memory store and routines; a person's signature on every step that changes something; every output a record people and programs can check.\n- Designed the kvkeys data model: every connector, skill and version of the department plugins in one Postgres schema (818 entities, 351 skills, 420 versions), replayed from history, every table and most columns commented, a 20-step CI gate before any merge.\n- Shipped kvkeys.com/#sql, Postgres running in the browser as the self-serve SQL console over that schema; built the jobs dataset behind this application: 293,891 postings from 13,750 Greenhouse, Lever and Ashby boards joined to the BLS occupation and employment tables.\n- Agents open the pull requests and a separate session reviews them before the merge, under rules I set: 685 merged across 39 repositories from 2 to 13 October 2026.\n\nEDUCATION\nUniversity of Washington | Seattle, WA\nBachelor of Arts, Economics; Minor, Entrepreneurship | Sep 2011 - Dec 2015\nGPA 3.5, Dean's List Scholar\n",
  "descriptionBody": "<h3>Skills</h3><ul><li><strong>Pipelines:</strong> dbt Core and Cloud, Airflow, Spark and PySpark, Hive, Databricks, AWS Glue and EMR, GitHub Actions CI/CD</li><li><strong>Warehouses:</strong> Snowflake, Redshift, Databricks, Azure Data Lake, Postgres; dimensional modeling (star schemas, slowly changing dimensions), semantic layers, data contracts, freshness SLAs and data quality KPIs</li><li><strong>Languages:</strong> SQL, Python (Pandas, PySpark, FastAPI, Streamlit), R</li><li><strong>Reporting:</strong> Superset, Tableau, Streamlit apps, SQL consoles in the browser (Postgres in the page); executive and cross-functional dashboards</li><li><strong>AI-assisted engineering:</strong> Claude Code, the Claude API and MCP connectors for pipeline, schema and CI work, with an independent review session before the merge; LLM-ready dbt metadata</li><li><strong>Cloud and tooling:</strong> AWS, Azure, Docker, GitHub Actions</li></ul><h3>Experience</h3><ul><li><strong>Senior Data Engineer</strong>, Coinbase (Remote, Mar 2024 - Present)<ul><li>Led the unified staking data model, a dimensional model across $18.9B in crypto assets on six networks, with Product Management, Data Science and leadership, using dbt Core and Cloud, Airflow and Snowflake; executive dashboards in Superset and Tableau.</li><li>Standardized the semantic layer across Coinbase: Python automation on the Claude API converted 1,600+ dbt, YAML and SQL files into LLM-ready metadata in the company's first data contracts rollout, so LLM-based analytics tools query trusted, documented datasets directly.</li><li>Built the org-wide observability platform in Streamlit (Python, SQL) that monitors 20+ pipeline health and data quality KPIs, with automated source freshness tests and schema validation across the warehouse, cutting the time to detect a failure.</li></ul></li><li><strong>Senior Data Engineer</strong>, Machina Labs (Chatsworth, CA, Feb 2023 - Mar 2024)<ul><li>First data engineering hire: built a robotics startup's Databricks warehouse from the ground up, PySpark ELT pipelines on Azure that bring robot telemetry, Salesforce and GraphQL feeds into a medallion lake for 40+ stakeholders, with CI/CD through GitHub Actions.</li><li>Built the FastAPI service over the Databricks APIs that ended direct cluster access for three internal teams; shipped the invoice management app and a bidirectional pricing calculator (Streamlit, PySpark, Docker on Azure) with write-back for Finance and Sales.</li></ul></li><li><strong>Data Scientist, Creator Monetization</strong>, TikTok (Los Angeles, CA, Dec 2021 - Feb 2023)<ul><li>Owned the Creator Monetization data stack end to end: 15+ production tables with dimensional modeling and 50+ Apache Hive SQL pipelines behind dashboards and cross-team analytics, with data contracts between event producers and downstream consumers.</li><li>Found and fixed a systematic error in the Branded Content revenue pipeline through SQL and lineage review, correcting projected revenue by $115M (+10%); built the Instacart add-to-cart ETL and reporting with anomaly detection, alerting and source freshness SLAs.</li></ul></li><li><strong>Business Intelligence Engineer II, Sponsored Products Advertising</strong>, Amazon (Seattle, WA, May 2020 - Dec 2021)<ul><li>Built and owned four Redshift ETL pipelines (80+ tasks) that publish the Weekly Business Review for the Targeting, Optimization, Recommendation and Budget teams: $450M+ in weekly ad revenue, zero SLA misses in 18 months; seven Tableau dashboards.</li><li>Owned the A/B test that launched Thematic placement sourcing (+$65MM annualized) and identified $8MM more in annualized revenue; standardized data models adopted across four teams.</li></ul></li><li><strong>Business Intelligence Engineer II, Private Brands</strong>, Amazon (Seattle, WA, Nov 2018 - May 2020)<ul><li>Automated product metrics in Redshift for the Accelerator program across 11 countries ($800M, 800% year over year), with slowly changing dimension tables for product attribute history; built the placement-level sales dashboard that steered marketing spend.</li><li>Traced sudden sales dips to bad actors ($300k opportunity cost) and shipped a real-time monitoring pipeline with data quality alerting and the takedown dashboard.</li></ul></li><li><strong>Custom Pricing Analyst</strong>, Amazon Web Services (Seattle, WA, Oct 2017 - Nov 2018)<ul><li>Designed and built the Enterprise Support analytical warehouse in Redshift (10M+ rows), a star-schema dimensional model that replaced manual billing workflows.</li><li>Owned the monthly billing pipeline for every AWS Enterprise Support customer: $55M a month in recurring billing, zero disputes; led two projects that added $1M in monthly AWS revenue.</li></ul></li><li><strong>Business Analyst (contract)</strong>, Amazon (Seattle, WA, Dec 2016 - Oct 2017)<ul><li>Built the Redshift pipelines and dashboards for contract-worker onboarding across 40+ countries and set the data governance standards for the HR system integrations; SQL, R and Python tooling for job-code mapping of the global contingent workforce.</li></ul></li><li><strong>Compensation Data Analyst (contract)</strong>, Microsoft (Redmond, WA, Sep 2016 - Dec 2016)<ul><li>R scripts that automated the manual revenue adjustment and bulk territory processes (60+ hours saved a month); quarterly audits of revenue, territories and quota.</li></ul></li><li><strong>Financial Analyst</strong>, Booking.com (Bellevue, WA, Jul 2015 - Sep 2015)<ul><li>Consolidated invoice reporting in SQL and Excel across two offices.</li></ul></li><li><strong>Financial Analyst, Retail</strong>, Amazon (Seattle, WA, Sep 2014 - Mar 2015)<ul><li>Optimized the SQL and automated the refresh of 20 Excel reports behind the Canada Retail weekly business review; customer returns and free replacements analyzed by reason code.</li></ul></li></ul><h3>Projects</h3><ul><li><strong>KVMACHINES, independent project</strong>, kvkeys.com/#sql, kvmachines.com (Seattle, WA, Jan 2026 - Present)<ul><li>Built to test whether agents can run a business on Anthropic's products: one Claude session per department with its own key, memory store and routines; a person's signature on every step that changes something; every output a record people and programs can check.</li><li>Designed the kvkeys data model: every connector, skill and version of the department plugins in one Postgres schema (818 entities, 351 skills, 420 versions), replayed from history, every table and most columns commented, a 20-step CI gate before any merge.</li><li>Shipped kvkeys.com/#sql, Postgres running in the browser as the self-serve SQL console over that schema; built the jobs dataset behind this application: 293,891 postings from 13,750 Greenhouse, Lever and Ashby boards joined to the BLS occupation and employment tables.</li><li>Agents open the pull requests and a separate session reviews them before the merge, under rules I set: 685 merged across 39 repositories from 2 to 13 October 2026.</li></ul></li></ul><h3>Education</h3><ul><li><strong>University of Washington</strong>, Seattle, WA: Bachelor of Arts, Economics; Minor, Entrepreneurship (Sep 2011 - Dec 2015); GPA 3.5, Dean's List Scholar</li></ul>",
  "descriptionBodyPlain": "ALEX ZHOUK\nSeattle, WA | https://www.linkedin.com/in/alexzhouk/ | https://kvkeys.com/#sql | https://kvmachines.com\nUS citizen. Based in Seattle. Open to Washington, California, and New York. Available to start on offer.\n\nSUMMARY\nSenior data engineer at Coinbase on dbt, Airflow and Snowflake: the unified staking data model, a semantic layer made LLM-ready with the Claude API, the observability platform behind 20+ KPIs. Data Engineer titles since 2023 after a Data Scientist role at TikTok and BI and analytics roles at Amazon; 10+ years in data roles. This year, KVMACHINES: an independent project to test whether agents can run a business on Anthropic's products, with a person's signature on every step that changes something (kvkeys.com/#sql).\n\nSKILLS\nPipelines: dbt Core and Cloud, Airflow, Spark and PySpark, Hive, Databricks, AWS Glue and EMR, GitHub Actions CI/CD\nWarehouses: Snowflake, Redshift, Databricks, Azure Data Lake, Postgres; dimensional modeling (star schemas, slowly changing dimensions), semantic layers, data contracts, freshness SLAs and data quality KPIs\nLanguages: SQL, Python (Pandas, PySpark, FastAPI, Streamlit), R\nReporting: Superset, Tableau, Streamlit apps, SQL consoles in the browser (Postgres in the page); executive and cross-functional dashboards\nAI-assisted engineering: Claude Code, the Claude API and MCP connectors for pipeline, schema and CI work, with an independent review session before the merge; LLM-ready dbt metadata\nCloud and tooling: AWS, Azure, Docker, GitHub Actions\n\nEXPERIENCE\n\nSenior Data Engineer\nCoinbase | Remote | Mar 2024 - Present\n- Led the unified staking data model, a dimensional model across $18.9B in crypto assets on six networks, with Product Management, Data Science and leadership, using dbt Core and Cloud, Airflow and Snowflake; executive dashboards in Superset and Tableau.\n- Standardized the semantic layer across Coinbase: Python automation on the Claude API converted 1,600+ dbt, YAML and SQL files into LLM-ready metadata in the company's first data contracts rollout, so LLM-based analytics tools query trusted, documented datasets directly.\n- Built the org-wide observability platform in Streamlit (Python, SQL) that monitors 20+ pipeline health and data quality KPIs, with automated source freshness tests and schema validation across the warehouse, cutting the time to detect a failure.\n\nSenior Data Engineer\nMachina Labs | Chatsworth, CA | Feb 2023 - Mar 2024\n- First data engineering hire: built a robotics startup's Databricks warehouse from the ground up, PySpark ELT pipelines on Azure that bring robot telemetry, Salesforce and GraphQL feeds into a medallion lake for 40+ stakeholders, with CI/CD through GitHub Actions.\n- Built the FastAPI service over the Databricks APIs that ended direct cluster access for three internal teams; shipped the invoice management app and a bidirectional pricing calculator (Streamlit, PySpark, Docker on Azure) with write-back for Finance and Sales.\n\nData Scientist, Creator Monetization\nTikTok | Los Angeles, CA | Dec 2021 - Feb 2023\n- Owned the Creator Monetization data stack end to end: 15+ production tables with dimensional modeling and 50+ Apache Hive SQL pipelines behind dashboards and cross-team analytics, with data contracts between event producers and downstream consumers.\n- Found and fixed a systematic error in the Branded Content revenue pipeline through SQL and lineage review, correcting projected revenue by $115M (+10%); built the Instacart add-to-cart ETL and reporting with anomaly detection, alerting and source freshness SLAs.\n\nBusiness Intelligence Engineer II, Sponsored Products Advertising\nAmazon | Seattle, WA | May 2020 - Dec 2021\n- Built and owned four Redshift ETL pipelines (80+ tasks) that publish the Weekly Business Review for the Targeting, Optimization, Recommendation and Budget teams: $450M+ in weekly ad revenue, zero SLA misses in 18 months; seven Tableau dashboards.\n- Owned the A/B test that launched Thematic placement sourcing (+$65MM annualized) and identified $8MM more in annualized revenue; standardized data models adopted across four teams.\n\nBusiness Intelligence Engineer II, Private Brands\nAmazon | Seattle, WA | Nov 2018 - May 2020\n- Automated product metrics in Redshift for the Accelerator program across 11 countries ($800M, 800% year over year), with slowly changing dimension tables for product attribute history; built the placement-level sales dashboard that steered marketing spend.\n- Traced sudden sales dips to bad actors ($300k opportunity cost) and shipped a real-time monitoring pipeline with data quality alerting and the takedown dashboard.\n\nCustom Pricing Analyst\nAmazon Web Services | Seattle, WA | Oct 2017 - Nov 2018\n- Designed and built the Enterprise Support analytical warehouse in Redshift (10M+ rows), a star-schema dimensional model that replaced manual billing workflows.\n- Owned the monthly billing pipeline for every AWS Enterprise Support customer: $55M a month in recurring billing, zero disputes; led two projects that added $1M in monthly AWS revenue.\n\nBusiness Analyst (contract)\nAmazon | Seattle, WA | Dec 2016 - Oct 2017\n- Built the Redshift pipelines and dashboards for contract-worker onboarding across 40+ countries and set the data governance standards for the HR system integrations; SQL, R and Python tooling for job-code mapping of the global contingent workforce.\n\nCompensation Data Analyst (contract)\nMicrosoft | Redmond, WA | Sep 2016 - Dec 2016\n- R scripts that automated the manual revenue adjustment and bulk territory processes (60+ hours saved a month); quarterly audits of revenue, territories and quota.\n\nFinancial Analyst\nBooking.com | Bellevue, WA | Jul 2015 - Sep 2015\n- Consolidated invoice reporting in SQL and Excel across two offices.\n\nFinancial Analyst, Retail\nAmazon | Seattle, WA | Sep 2014 - Mar 2015\n- Optimized the SQL and automated the refresh of 20 Excel reports behind the Canada Retail weekly business review; customer returns and free replacements analyzed by reason code.\n\nPROJECTS\n\nKVMACHINES, independent project\nkvkeys.com/#sql, kvmachines.com | Seattle, WA | Jan 2026 - Present\n- Built to test whether agents can run a business on Anthropic's products: one Claude session per department with its own key, memory store and routines; a person's signature on every step that changes something; every output a record people and programs can check.\n- Designed the kvkeys data model: every connector, skill and version of the department plugins in one Postgres schema (818 entities, 351 skills, 420 versions), replayed from history, every table and most columns commented, a 20-step CI gate before any merge.\n- Shipped kvkeys.com/#sql, Postgres running in the browser as the self-serve SQL console over that schema; built the jobs dataset behind this application: 293,891 postings from 13,750 Greenhouse, Lever and Ashby boards joined to the BLS occupation and employment tables.\n- Agents open the pull requests and a separate session reviews them before the merge, under rules I set: 685 merged across 39 repositories from 2 to 13 October 2026.\n\nEDUCATION\nUniversity of Washington | Seattle, WA\nBachelor of Arts, Economics; Minor, Entrepreneurship | Sep 2011 - Dec 2015\nGPA 3.5, Dean's List Scholar\n",
  "lists": [
    {
      "text": "Skills",
      "content": "<li><strong>Pipelines:</strong> dbt Core and Cloud, Airflow, Spark and PySpark, Hive, Databricks, AWS Glue and EMR, GitHub Actions CI/CD</li><li><strong>Warehouses:</strong> Snowflake, Redshift, Databricks, Azure Data Lake, Postgres; dimensional modeling (star schemas, slowly changing dimensions), semantic layers, data contracts, freshness SLAs and data quality KPIs</li><li><strong>Languages:</strong> SQL, Python (Pandas, PySpark, FastAPI, Streamlit), R</li><li><strong>Reporting:</strong> Superset, Tableau, Streamlit apps, SQL consoles in the browser (Postgres in the page); executive and cross-functional dashboards</li><li><strong>AI-assisted engineering:</strong> Claude Code, the Claude API and MCP connectors for pipeline, schema and CI work, with an independent review session before the merge; LLM-ready dbt metadata</li><li><strong>Cloud and tooling:</strong> AWS, Azure, Docker, GitHub Actions</li>"
    },
    {
      "text": "Experience",
      "content": "<li><strong>Senior Data Engineer</strong>, Coinbase (Remote, Mar 2024 - Present)<ul><li>Led the unified staking data model, a dimensional model across $18.9B in crypto assets on six networks, with Product Management, Data Science and leadership, using dbt Core and Cloud, Airflow and Snowflake; executive dashboards in Superset and Tableau.</li><li>Standardized the semantic layer across Coinbase: Python automation on the Claude API converted 1,600+ dbt, YAML and SQL files into LLM-ready metadata in the company's first data contracts rollout, so LLM-based analytics tools query trusted, documented datasets directly.</li><li>Built the org-wide observability platform in Streamlit (Python, SQL) that monitors 20+ pipeline health and data quality KPIs, with automated source freshness tests and schema validation across the warehouse, cutting the time to detect a failure.</li></ul></li><li><strong>Senior Data Engineer</strong>, Machina Labs (Chatsworth, CA, Feb 2023 - Mar 2024)<ul><li>First data engineering hire: built a robotics startup's Databricks warehouse from the ground up, PySpark ELT pipelines on Azure that bring robot telemetry, Salesforce and GraphQL feeds into a medallion lake for 40+ stakeholders, with CI/CD through GitHub Actions.</li><li>Built the FastAPI service over the Databricks APIs that ended direct cluster access for three internal teams; shipped the invoice management app and a bidirectional pricing calculator (Streamlit, PySpark, Docker on Azure) with write-back for Finance and Sales.</li></ul></li><li><strong>Data Scientist, Creator Monetization</strong>, TikTok (Los Angeles, CA, Dec 2021 - Feb 2023)<ul><li>Owned the Creator Monetization data stack end to end: 15+ production tables with dimensional modeling and 50+ Apache Hive SQL pipelines behind dashboards and cross-team analytics, with data contracts between event producers and downstream consumers.</li><li>Found and fixed a systematic error in the Branded Content revenue pipeline through SQL and lineage review, correcting projected revenue by $115M (+10%); built the Instacart add-to-cart ETL and reporting with anomaly detection, alerting and source freshness SLAs.</li></ul></li><li><strong>Business Intelligence Engineer II, Sponsored Products Advertising</strong>, Amazon (Seattle, WA, May 2020 - Dec 2021)<ul><li>Built and owned four Redshift ETL pipelines (80+ tasks) that publish the Weekly Business Review for the Targeting, Optimization, Recommendation and Budget teams: $450M+ in weekly ad revenue, zero SLA misses in 18 months; seven Tableau dashboards.</li><li>Owned the A/B test that launched Thematic placement sourcing (+$65MM annualized) and identified $8MM more in annualized revenue; standardized data models adopted across four teams.</li></ul></li><li><strong>Business Intelligence Engineer II, Private Brands</strong>, Amazon (Seattle, WA, Nov 2018 - May 2020)<ul><li>Automated product metrics in Redshift for the Accelerator program across 11 countries ($800M, 800% year over year), with slowly changing dimension tables for product attribute history; built the placement-level sales dashboard that steered marketing spend.</li><li>Traced sudden sales dips to bad actors ($300k opportunity cost) and shipped a real-time monitoring pipeline with data quality alerting and the takedown dashboard.</li></ul></li><li><strong>Custom Pricing Analyst</strong>, Amazon Web Services (Seattle, WA, Oct 2017 - Nov 2018)<ul><li>Designed and built the Enterprise Support analytical warehouse in Redshift (10M+ rows), a star-schema dimensional model that replaced manual billing workflows.</li><li>Owned the monthly billing pipeline for every AWS Enterprise Support customer: $55M a month in recurring billing, zero disputes; led two projects that added $1M in monthly AWS revenue.</li></ul></li><li><strong>Business Analyst (contract)</strong>, Amazon (Seattle, WA, Dec 2016 - Oct 2017)<ul><li>Built the Redshift pipelines and dashboards for contract-worker onboarding across 40+ countries and set the data governance standards for the HR system integrations; SQL, R and Python tooling for job-code mapping of the global contingent workforce.</li></ul></li><li><strong>Compensation Data Analyst (contract)</strong>, Microsoft (Redmond, WA, Sep 2016 - Dec 2016)<ul><li>R scripts that automated the manual revenue adjustment and bulk territory processes (60+ hours saved a month); quarterly audits of revenue, territories and quota.</li></ul></li><li><strong>Financial Analyst</strong>, Booking.com (Bellevue, WA, Jul 2015 - Sep 2015)<ul><li>Consolidated invoice reporting in SQL and Excel across two offices.</li></ul></li><li><strong>Financial Analyst, Retail</strong>, Amazon (Seattle, WA, Sep 2014 - Mar 2015)<ul><li>Optimized the SQL and automated the refresh of 20 Excel reports behind the Canada Retail weekly business review; customer returns and free replacements analyzed by reason code.</li></ul></li>"
    },
    {
      "text": "Projects",
      "content": "<li><strong>KVMACHINES, independent project</strong>, kvkeys.com/#sql, kvmachines.com (Seattle, WA, Jan 2026 - Present)<ul><li>Built to test whether agents can run a business on Anthropic's products: one Claude session per department with its own key, memory store and routines; a person's signature on every step that changes something; every output a record people and programs can check.</li><li>Designed the kvkeys data model: every connector, skill and version of the department plugins in one Postgres schema (818 entities, 351 skills, 420 versions), replayed from history, every table and most columns commented, a 20-step CI gate before any merge.</li><li>Shipped kvkeys.com/#sql, Postgres running in the browser as the self-serve SQL console over that schema; built the jobs dataset behind this application: 293,891 postings from 13,750 Greenhouse, Lever and Ashby boards joined to the BLS occupation and employment tables.</li><li>Agents open the pull requests and a separate session reviews them before the merge, under rules I set: 685 merged across 39 repositories from 2 to 13 October 2026.</li></ul></li>"
    },
    {
      "text": "Education",
      "content": "<li><strong>University of Washington</strong>, Seattle, WA: Bachelor of Arts, Economics; Minor, Entrepreneurship (Sep 2011 - Dec 2015); GPA 3.5, Dean's List Scholar</li>"
    }
  ],
  "additional": "<p>US citizen. Based in Seattle. Open to Washington, California, and New York. Available to start on offer.</p>",
  "additionalPlain": "US citizen. Based in Seattle. Open to Washington, California, and New York. Available to start on offer.",
  "hostedUrl": "https://alexzhouk.com/",
  "applyUrl": "https://www.linkedin.com/in/alexzhouk/",
  "workplaceType": "unspecified"
}

{
  "apiVersion": "1",
  "jobs": [
    {
      "title": "Senior Data Engineer",
      "location": "Seattle, WA",
      "secondaryLocations": [
        {
          "location": "Washington",
          "address": {
            "addressRegion": "Washington",
            "addressCountry": "USA"
          }
        },
        {
          "location": "California",
          "address": {
            "addressRegion": "California",
            "addressCountry": "USA"
          }
        },
        {
          "location": "New York",
          "address": {
            "addressRegion": "New York",
            "addressCountry": "USA"
          }
        }
      ],
      "department": null,
      "team": null,
      "isListed": true,
      "descriptionHtml": "<p>Senior data engineer at Coinbase on dbt, Airflow and Snowflake: the unified staking data model, a semantic layer made LLM-ready with the Claude API, the observability platform behind 20+ KPIs. Data Engineer titles since 2023 after a Data Scientist role at TikTok and BI and analytics roles at Amazon; 10+ years in data roles. This year, KVMACHINES: an independent project to test whether agents can run a business on Anthropic's products, with a person's signature on every step that changes something (kvkeys.com/#sql).</p><h3>Skills</h3><ul><li><strong>Pipelines:</strong> dbt Core and Cloud, Airflow, Spark and PySpark, Hive, Databricks, AWS Glue and EMR, GitHub Actions CI/CD</li><li><strong>Warehouses:</strong> Snowflake, Redshift, Databricks, Azure Data Lake, Postgres; dimensional modeling (star schemas, slowly changing dimensions), semantic layers, data contracts, freshness SLAs and data quality KPIs</li><li><strong>Languages:</strong> SQL, Python (Pandas, PySpark, FastAPI, Streamlit), R</li><li><strong>Reporting:</strong> Superset, Tableau, Streamlit apps, SQL consoles in the browser (Postgres in the page); executive and cross-functional dashboards</li><li><strong>AI-assisted engineering:</strong> Claude Code, the Claude API and MCP connectors for pipeline, schema and CI work, with an independent review session before the merge; LLM-ready dbt metadata</li><li><strong>Cloud and tooling:</strong> AWS, Azure, Docker, GitHub Actions</li></ul><h3>Experience</h3><ul><li><strong>Senior Data Engineer</strong>, Coinbase (Remote, Mar 2024 - Present)<ul><li>Led the unified staking data model, a dimensional model across $18.9B in crypto assets on six networks, with Product Management, Data Science and leadership, using dbt Core and Cloud, Airflow and Snowflake; executive dashboards in Superset and Tableau.</li><li>Standardized the semantic layer across Coinbase: Python automation on the Claude API converted 1,600+ dbt, YAML and SQL files into LLM-ready metadata in the company's first data contracts rollout, so LLM-based analytics tools query trusted, documented datasets directly.</li><li>Built the org-wide observability platform in Streamlit (Python, SQL) that monitors 20+ pipeline health and data quality KPIs, with automated source freshness tests and schema validation across the warehouse, cutting the time to detect a failure.</li></ul></li><li><strong>Senior Data Engineer</strong>, Machina Labs (Chatsworth, CA, Feb 2023 - Mar 2024)<ul><li>First data engineering hire: built a robotics startup's Databricks warehouse from the ground up, PySpark ELT pipelines on Azure that bring robot telemetry, Salesforce and GraphQL feeds into a medallion lake for 40+ stakeholders, with CI/CD through GitHub Actions.</li><li>Built the FastAPI service over the Databricks APIs that ended direct cluster access for three internal teams; shipped the invoice management app and a bidirectional pricing calculator (Streamlit, PySpark, Docker on Azure) with write-back for Finance and Sales.</li></ul></li><li><strong>Data Scientist, Creator Monetization</strong>, TikTok (Los Angeles, CA, Dec 2021 - Feb 2023)<ul><li>Owned the Creator Monetization data stack end to end: 15+ production tables with dimensional modeling and 50+ Apache Hive SQL pipelines behind dashboards and cross-team analytics, with data contracts between event producers and downstream consumers.</li><li>Found and fixed a systematic error in the Branded Content revenue pipeline through SQL and lineage review, correcting projected revenue by $115M (+10%); built the Instacart add-to-cart ETL and reporting with anomaly detection, alerting and source freshness SLAs.</li></ul></li><li><strong>Business Intelligence Engineer II, Sponsored Products Advertising</strong>, Amazon (Seattle, WA, May 2020 - Dec 2021)<ul><li>Built and owned four Redshift ETL pipelines (80+ tasks) that publish the Weekly Business Review for the Targeting, Optimization, Recommendation and Budget teams: $450M+ in weekly ad revenue, zero SLA misses in 18 months; seven Tableau dashboards.</li><li>Owned the A/B test that launched Thematic placement sourcing (+$65MM annualized) and identified $8MM more in annualized revenue; standardized data models adopted across four teams.</li></ul></li><li><strong>Business Intelligence Engineer II, Private Brands</strong>, Amazon (Seattle, WA, Nov 2018 - May 2020)<ul><li>Automated product metrics in Redshift for the Accelerator program across 11 countries ($800M, 800% year over year), with slowly changing dimension tables for product attribute history; built the placement-level sales dashboard that steered marketing spend.</li><li>Traced sudden sales dips to bad actors ($300k opportunity cost) and shipped a real-time monitoring pipeline with data quality alerting and the takedown dashboard.</li></ul></li><li><strong>Custom Pricing Analyst</strong>, Amazon Web Services (Seattle, WA, Oct 2017 - Nov 2018)<ul><li>Designed and built the Enterprise Support analytical warehouse in Redshift (10M+ rows), a star-schema dimensional model that replaced manual billing workflows.</li><li>Owned the monthly billing pipeline for every AWS Enterprise Support customer: $55M a month in recurring billing, zero disputes; led two projects that added $1M in monthly AWS revenue.</li></ul></li><li><strong>Business Analyst (contract)</strong>, Amazon (Seattle, WA, Dec 2016 - Oct 2017)<ul><li>Built the Redshift pipelines and dashboards for contract-worker onboarding across 40+ countries and set the data governance standards for the HR system integrations; SQL, R and Python tooling for job-code mapping of the global contingent workforce.</li></ul></li><li><strong>Compensation Data Analyst (contract)</strong>, Microsoft (Redmond, WA, Sep 2016 - Dec 2016)<ul><li>R scripts that automated the manual revenue adjustment and bulk territory processes (60+ hours saved a month); quarterly audits of revenue, territories and quota.</li></ul></li><li><strong>Financial Analyst</strong>, Booking.com (Bellevue, WA, Jul 2015 - Sep 2015)<ul><li>Consolidated invoice reporting in SQL and Excel across two offices.</li></ul></li><li><strong>Financial Analyst, Retail</strong>, Amazon (Seattle, WA, Sep 2014 - Mar 2015)<ul><li>Optimized the SQL and automated the refresh of 20 Excel reports behind the Canada Retail weekly business review; customer returns and free replacements analyzed by reason code.</li></ul></li></ul><h3>Projects</h3><ul><li><strong>KVMACHINES, independent project</strong>, kvkeys.com/#sql, kvmachines.com (Seattle, WA, Jan 2026 - Present)<ul><li>Built to test whether agents can run a business on Anthropic's products: one Claude session per department with its own key, memory store and routines; a person's signature on every step that changes something; every output a record people and programs can check.</li><li>Designed the kvkeys data model: every connector, skill and version of the department plugins in one Postgres schema (818 entities, 351 skills, 420 versions), replayed from history, every table and most columns commented, a 20-step CI gate before any merge.</li><li>Shipped kvkeys.com/#sql, Postgres running in the browser as the self-serve SQL console over that schema; built the jobs dataset behind this application: 293,891 postings from 13,750 Greenhouse, Lever and Ashby boards joined to the BLS occupation and employment tables.</li><li>Agents open the pull requests and a separate session reviews them before the merge, under rules I set: 685 merged across 39 repositories from 2 to 13 October 2026.</li></ul></li></ul><h3>Education</h3><ul><li><strong>University of Washington</strong>, Seattle, WA: Bachelor of Arts, Economics; Minor, Entrepreneurship (Sep 2011 - Dec 2015); GPA 3.5, Dean's List Scholar</li></ul><p>US citizen. Based in Seattle. Open to Washington, California, and New York. Available to start on offer.</p>",
      "descriptionPlain": "ALEX ZHOUK\nSeattle, WA | https://www.linkedin.com/in/alexzhouk/ | https://kvkeys.com/#sql | https://kvmachines.com\nUS citizen. Based in Seattle. Open to Washington, California, and New York. Available to start on offer.\n\nSUMMARY\nSenior data engineer at Coinbase on dbt, Airflow and Snowflake: the unified staking data model, a semantic layer made LLM-ready with the Claude API, the observability platform behind 20+ KPIs. Data Engineer titles since 2023 after a Data Scientist role at TikTok and BI and analytics roles at Amazon; 10+ years in data roles. This year, KVMACHINES: an independent project to test whether agents can run a business on Anthropic's products, with a person's signature on every step that changes something (kvkeys.com/#sql).\n\nSKILLS\nPipelines: dbt Core and Cloud, Airflow, Spark and PySpark, Hive, Databricks, AWS Glue and EMR, GitHub Actions CI/CD\nWarehouses: Snowflake, Redshift, Databricks, Azure Data Lake, Postgres; dimensional modeling (star schemas, slowly changing dimensions), semantic layers, data contracts, freshness SLAs and data quality KPIs\nLanguages: SQL, Python (Pandas, PySpark, FastAPI, Streamlit), R\nReporting: Superset, Tableau, Streamlit apps, SQL consoles in the browser (Postgres in the page); executive and cross-functional dashboards\nAI-assisted engineering: Claude Code, the Claude API and MCP connectors for pipeline, schema and CI work, with an independent review session before the merge; LLM-ready dbt metadata\nCloud and tooling: AWS, Azure, Docker, GitHub Actions\n\nEXPERIENCE\n\nSenior Data Engineer\nCoinbase | Remote | Mar 2024 - Present\n- Led the unified staking data model, a dimensional model across $18.9B in crypto assets on six networks, with Product Management, Data Science and leadership, using dbt Core and Cloud, Airflow and Snowflake; executive dashboards in Superset and Tableau.\n- Standardized the semantic layer across Coinbase: Python automation on the Claude API converted 1,600+ dbt, YAML and SQL files into LLM-ready metadata in the company's first data contracts rollout, so LLM-based analytics tools query trusted, documented datasets directly.\n- Built the org-wide observability platform in Streamlit (Python, SQL) that monitors 20+ pipeline health and data quality KPIs, with automated source freshness tests and schema validation across the warehouse, cutting the time to detect a failure.\n\nSenior Data Engineer\nMachina Labs | Chatsworth, CA | Feb 2023 - Mar 2024\n- First data engineering hire: built a robotics startup's Databricks warehouse from the ground up, PySpark ELT pipelines on Azure that bring robot telemetry, Salesforce and GraphQL feeds into a medallion lake for 40+ stakeholders, with CI/CD through GitHub Actions.\n- Built the FastAPI service over the Databricks APIs that ended direct cluster access for three internal teams; shipped the invoice management app and a bidirectional pricing calculator (Streamlit, PySpark, Docker on Azure) with write-back for Finance and Sales.\n\nData Scientist, Creator Monetization\nTikTok | Los Angeles, CA | Dec 2021 - Feb 2023\n- Owned the Creator Monetization data stack end to end: 15+ production tables with dimensional modeling and 50+ Apache Hive SQL pipelines behind dashboards and cross-team analytics, with data contracts between event producers and downstream consumers.\n- Found and fixed a systematic error in the Branded Content revenue pipeline through SQL and lineage review, correcting projected revenue by $115M (+10%); built the Instacart add-to-cart ETL and reporting with anomaly detection, alerting and source freshness SLAs.\n\nBusiness Intelligence Engineer II, Sponsored Products Advertising\nAmazon | Seattle, WA | May 2020 - Dec 2021\n- Built and owned four Redshift ETL pipelines (80+ tasks) that publish the Weekly Business Review for the Targeting, Optimization, Recommendation and Budget teams: $450M+ in weekly ad revenue, zero SLA misses in 18 months; seven Tableau dashboards.\n- Owned the A/B test that launched Thematic placement sourcing (+$65MM annualized) and identified $8MM more in annualized revenue; standardized data models adopted across four teams.\n\nBusiness Intelligence Engineer II, Private Brands\nAmazon | Seattle, WA | Nov 2018 - May 2020\n- Automated product metrics in Redshift for the Accelerator program across 11 countries ($800M, 800% year over year), with slowly changing dimension tables for product attribute history; built the placement-level sales dashboard that steered marketing spend.\n- Traced sudden sales dips to bad actors ($300k opportunity cost) and shipped a real-time monitoring pipeline with data quality alerting and the takedown dashboard.\n\nCustom Pricing Analyst\nAmazon Web Services | Seattle, WA | Oct 2017 - Nov 2018\n- Designed and built the Enterprise Support analytical warehouse in Redshift (10M+ rows), a star-schema dimensional model that replaced manual billing workflows.\n- Owned the monthly billing pipeline for every AWS Enterprise Support customer: $55M a month in recurring billing, zero disputes; led two projects that added $1M in monthly AWS revenue.\n\nBusiness Analyst (contract)\nAmazon | Seattle, WA | Dec 2016 - Oct 2017\n- Built the Redshift pipelines and dashboards for contract-worker onboarding across 40+ countries and set the data governance standards for the HR system integrations; SQL, R and Python tooling for job-code mapping of the global contingent workforce.\n\nCompensation Data Analyst (contract)\nMicrosoft | Redmond, WA | Sep 2016 - Dec 2016\n- R scripts that automated the manual revenue adjustment and bulk territory processes (60+ hours saved a month); quarterly audits of revenue, territories and quota.\n\nFinancial Analyst\nBooking.com | Bellevue, WA | Jul 2015 - Sep 2015\n- Consolidated invoice reporting in SQL and Excel across two offices.\n\nFinancial Analyst, Retail\nAmazon | Seattle, WA | Sep 2014 - Mar 2015\n- Optimized the SQL and automated the refresh of 20 Excel reports behind the Canada Retail weekly business review; customer returns and free replacements analyzed by reason code.\n\nPROJECTS\n\nKVMACHINES, independent project\nkvkeys.com/#sql, kvmachines.com | Seattle, WA | Jan 2026 - Present\n- Built to test whether agents can run a business on Anthropic's products: one Claude session per department with its own key, memory store and routines; a person's signature on every step that changes something; every output a record people and programs can check.\n- Designed the kvkeys data model: every connector, skill and version of the department plugins in one Postgres schema (818 entities, 351 skills, 420 versions), replayed from history, every table and most columns commented, a 20-step CI gate before any merge.\n- Shipped kvkeys.com/#sql, Postgres running in the browser as the self-serve SQL console over that schema; built the jobs dataset behind this application: 293,891 postings from 13,750 Greenhouse, Lever and Ashby boards joined to the BLS occupation and employment tables.\n- Agents open the pull requests and a separate session reviews them before the merge, under rules I set: 685 merged across 39 repositories from 2 to 13 October 2026.\n\nEDUCATION\nUniversity of Washington | Seattle, WA\nBachelor of Arts, Economics; Minor, Entrepreneurship | Sep 2011 - Dec 2015\nGPA 3.5, Dean's List Scholar\n",
      "publishedAt": "2026-10-13T00:00:00.000+00:00",
      "address": {
        "postalAddress": {
          "addressLocality": "Seattle",
          "addressRegion": "Washington",
          "addressCountry": "USA"
        }
      },
      "jobUrl": "https://alexzhouk.com/",
      "applyUrl": "https://www.linkedin.com/in/alexzhouk/"
    }
  ]
}